On February 9, 2015, Tristan wrote:
Note, one gna test fails, for both .exe's, when using 4a8b24 :
GNA test failed ( bug05)
This is the vcd/fst dump test. Is it the vcd or the fst test
that fails ? How does it fail ?
Failure log:
"
" GNA dir bug05:
" analyze repro.vhdl
" elaborate and simulate tb --fst=tb.fst
" c:\Ghdl\ghdl-0.33dev-batch\bin\ghdl.exe:error: unknown option
'--fst=tb.fst', try --help
"
--------------------
Is is due to ghdlfilter ? In that case, we could either
set the encoding ( I have to look at the GNAT doc to
understand that issue), or we can modify ghdlfilter
to avoid that issue.
The substitution *only* occurs when running the PowerShell build script.
Running ghdlfilter from the original batch file preserves the copyright
symbol.
That's why I suspected the PowerShell pipe mechanism in my original
post comparing the .bat vs. .ps1 build results:
Difference noted, probably not worth much effort to fix:
- Copyright symbols in the 2008 ieee library source comments
have been replaced by "?" after preprocessing
( Some sort of PowerShell pipe character set issue? )
-Brian
_______________________________________________
Ghdl-discuss mailing list
[email protected]
https://mail.gna.org/listinfo/ghdl-discuss
